Finite element(FEM) modelling of Wire-Arc-Additive-Manufacturing process

Authors: Filippo Montevecchia, Giuseppe Venturinia, Antonio Scippaa , Gianni Campatellia

Keywords: Welding; Finite element method (FEM); Additive manufacturing;

Abstract: Wire-Arc-Additive-Manufacturing (WAAM) is an Additive-Manufacturing (AM) process, allowing to produce metal components layer by layer by means of Gas-Metal-Arc-Welding (GMAW) technology. The advantages of this technology are the capability to create large parts with a higher deposition rate with respect to other AM technologies. Despite these great benefits, WAAM components are affected by severe distortions and residual stresses issues. Finite element process simulation provides an efficient way to study mitigation strategies for such issues. In this paper, a WAAM modelling strategy is proposed based on a novel heat source model that takes into account the actual power distribution between filler and base materials. In order to prove the effectiveness of proposed modelling, an experimental validation is provided by comparing the measured distortions of a WAAM tests-case with the simulated ones, highlighting the accuracy of proposed model.

Purpose: Simulate the WAAM process to test the effect of different deposition patterns on residual stresses field, optimizing the process

Methodology: The heat transfer from the arc to the molten pool is simulated using a heat source model, which prescribes a heat generation per unit volume in the molten pool region. Material deposition is taken into account by means of specific elements activation algorithms. In this model the heat input is delivered over a moving double ellipsoid region according to a Gaussian distribution

Findings: Proposed process modelling allows to accurately simulate the WAAM process, without the need to perform time-consuming tuning operations to identify heat source parameters.

Link: https://www.sciencedirect.com/science/article/pii/S2212827116309131

Limitations: There are some inaccuracies infact, percentage errors on the maximum displacements points are 2% and 26%. Despite the latter result may seem inaccurate, it should be considered that material data have been derived from literature, hence actual material behavior could be different from the model one.

Benefits: Proposed modelling results are in general agreement with the experimental ones, allowing to achieve an higher accuracy with respect to the traditional technique